Polycyclic Aromatic Hydrocarbons (PAHs) in Surface Soils in … · 2020-05-01 · Polycyclic...
Transcript of Polycyclic Aromatic Hydrocarbons (PAHs) in Surface Soils in … · 2020-05-01 · Polycyclic...
Polycyclic Aromatic Hydrocarbons (PAHs)in Surface Soils in Western New York
Follow-up Report for New York State Soils
Technical Report
EPRI Project Manager A. Coleman
EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA 800.313.3774 • 650.855.2121 • [email protected] • www.epri.com
Polycyclic Aromatic Hydrocarbons (PAHs) in Surface Soils in Western New York Follow-up Report for New York State Soils
1005296
Final Report, October 2003
Cosponsors New York State Electric & Gas Corporation Rochester Gas and Electric Corporation Public Service Enterprise Group Services, LLP
DISCLAIMER OF WARRANTIES AND LIMITATION OF LIABILITIES
THIS DOCUMENT WAS PREPARED BY THE ORGANIZATION(S) NAMED BELOW AS AN ACCOUNT OF WORK SPONSORED OR COSPONSORED BY THE ELECTRIC POWER RESEARCH INSTITUTE, INC. (EPRI). NEITHER EPRI, ANY MEMBER OF EPRI, ANY COSPONSOR, THE ORGANIZATION(S) BELOW, NOR ANY PERSON ACTING ON BEHALF OF ANY OF THEM:
(A) MAKES ANY WARRANTY OR REPRESENTATION WHATSOEVER, EXPRESS OR IMPLIED, (I) WITH RESPECT TO THE USE OF ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT, INCLUDING MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE, OR (II) THAT SUCH USE DOES NOT INFRINGE ON OR INTERFERE WITH PRIVATELY OWNED RIGHTS, INCLUDING ANY PARTY'S INTELLECTUAL PROPERTY, OR (III) THAT THIS DOCUMENT IS SUITABLE TO ANY PARTICULAR USER'S CIRCUMSTANCE; OR
(B) ASSUMES RESPONSIBILITY FOR ANY DAMAGES OR OTHER LIABILITY WHATSOEVER (INCLUDING ANY CONSEQUENTIAL DAMAGES, EVEN IF EPRI OR ANY EPRI REPRESENTATIVE HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES) RESULTING FROM YOUR SELECTION OR USE OF THIS DOCUMENT OR ANY INFORMATION, APPARATUS, METHOD, PROCESS, OR SIMILAR ITEM DISCLOSED IN THIS DOCUMENT.
ORGANIZATION(S) THAT PREPARED THIS DOCUMENT
META Environmental, Inc.
ORDERING INFORMATION
Requests for copies of this report should be directed to EPRI Orders and Conferences, 1355 Willow Way, Suite 278, Concord, CA 94520, (800) 313-3774, press 2 or internally x5379, (925) 609-9169, (925) 609-1310 (fax).
Electric Power Research Institute and EPRI are registered service marks of the Electric Power Research Institute, Inc. EPRI. ELECTRIFY THE WORLD is a service mark of the Electric Power Research Institute, Inc.
Copyright © 2003 Electric Power Research Institute, Inc. All rights reserved.
iii
CITATIONS
This report was prepared by
META Environmental, Inc. 49 Clarendon Street Watertown, MA 02472
Principal Investigators D. Mauro P. De Clercq
This report was prepared for
EPRI 3412 Hillview Avenue Palo Alto, California 94304 and
New York State Electric & Gas Corporation P.O. Box 5224 Binghamton, NY 13902
Project Managers T. Blazicek J. Simone
and
Rochester Gas and Electric Corporation 89 East Avenue Rochester, NY 14649
Project Managers K. Sahler K. Hylton
and
iv
Public Service Enterprise Group Services, LLP 80 Park Plaza Newark, NJ 07101
Project Manager W. Max
This report describes research sponsored by EPRI, New York State Electric & Gas Corporation, Rochester Gas and Electric Corporation, and Public Service Enterprise Group Services, LLP.
The report is a corporate document that should be cited in the literature in the following manner:
Polycyclic Aromatic Hydrocarbons (PAHs) in Surface Soils in Western New York: Follow-up Report for New York State Soils, EPRI, Palo Alto, CA, New York State Electric & Gas Corporation, Binghamton, NY, Rochester Gas and Electric Corporation, Rochester, NY, and Public Service Enterprise Group Services, LLP, Newark, NJ: 2003. 1005296.
v
PRODUCT DESCRIPTION
Natural and anthropogenic sources of polycyclic aromatic hydrocarbons (PAHs) in the environment are quite diverse due to many different factors of human activity or inactivity in an area. Consequently, PAHs are found in many samples of soil, water, and biota throughout the world. This report provides an update of ongoing EPRI research performed to examine the distribution of PAH concentrations in surface soils of populated areas, with emphasis here on recently collected data in western New York State. This specific project completes the data set of EPRI’s sampling of background soils in western and northwestern New York State (report 1005295).
Results & Findings A total of 30 sites from two municipalities (Canandaigua and Rochester) in western New York State were sampled as part of an ongoing EPRI background study of PAHs in surface soil. Each municipality had a population of at least 10,000 people and a population density of at least 1000 people per square mile. At each site, two locations were sampled from 0 – 2.54 cm below grade and from 2.54 – 15 cm below grade, after removing surface vegetation, if present. At all sites, samples were composited to generate one sample representative of 0 – 15 cm below grade.
The patterns of PAHs seen in the samples were consistent with those in surface soil samples containing PAHs from pyrogenic sources that have been subjected to environmental weathering. To compare the concentrations of individual PAHs for all of the sites, summary statistics were performed with not-detected results equal to one half of the detection limit. The results for the individual PAHs were generally lognormally distributed. Benzo(a)pyrene concentrations ranged from a non-detectable 7.5 – 4740 µg/kg, with a median concentration of 431 µg/kg. The upper quartile concentration was 1040 µg/kg, while the upper 95th percentile concentration was 2770 µg/kg. These concentrations appear to be consistent with literature-reported levels for anthropogenic background in small- to medium-sized residential, commercial, and light industrial areas, but notably lower than background observed in more highly populated cities and commercial/industrial areas.
Challenges & Objectives At many sites, it becomes very important to be able to determine where the site-specific contamination ends and where the local background PAHs begin. EPRI’s primary objectives in the study of background PAHs in Illinois and western New York were as follows: • To collect, composite (if appropriate), analyze, and interpret the results of surface soil
samples from commercial/industrial, residential, municipal, and rural locations.
• To compile the data, determine their distribution and statistics, and correlate the PAH concentrations and patterns to site conditions or histories, when possible.
vi
• To generate detailed information on typical PAH levels found at different locations as an aid in establishing appropriate and realistic cleanup levels for surface soil PAHs at sites undergoing remedial actions.
Applications, Values & Use The frequent detection of PAHs in surface soil and sediment samples is of particular importance for environmental investigations and cleanups, because PAH concentrations often define the extent of contamination and influence the estimated risk from contamination at a variety of sites. This EPRI Background PAH Study was designed to generate an internally consistent, rugged set of PAH concentrations in surface soils from numerous locations covering a range of land uses, for application in site remediation plans.
EPRI Perspective The technical and cost considerations that guide remediation are diverse, and during remediation, background conditions must be considered. Although there are many literature references concerning the presence of PAHs in environmental media, including surface soils, a review of those references indicated that there is a lack of consistent data on background distributions and concentrations of PAHs in surface soils. Indeed, there is no consistent definition of “background” in a regulatory sense. This study—cosponsored by EPRI, New York State Electric & Gas Corporation, Rochester Gas and Electric Corporation, and Public Service Enterprise Group Services—describes the nature of the distribution of PAH concentrations in populated areas. It identifies factors that affect distribution, such as type of setting (urban or rural), sample depth, soil type, and proximity to known sources. Finally, it examines, in a systematic manner, chemicals or chemical patterns that provide indications of PAH sources detected in background samples. Such information will prove invaluable in remediation of PAH-contaminated soil and in further differentiating between petrogenic and pyrogenic wastes.
Approach The study objectives were achieved through a program of sample collection and analysis based on random sampling of a large number of sites with no known point sources of PAH-containing wastes. Sites were selected from urban areas and included utility-owned land, parks, forested areas, roadside rights of way, fields, public properties, and commercial properties. Selected samples were composited in the laboratory prior to analysis, with only two and rarely three samples combined into one composite sample. Samples were analyzed for PAHs and total organic carbon (TOC).
Keywords Soils Contamination Hydrocarbons Manufactured Gas Plant Sites Coal Tar Polycyclic Aromatic Hydrocarbons
vii
ABSTRACT
The frequent detection of polycyclic aromatic hydrocarbons (PAHs) in surface soil and sediment samples is of particular importance for environmental investigations and cleanups, because PAH concentrations often define the extent of contamination and influence the estimated risk from contamination at a variety of sites. A total of 30 sites from two municipalities (Canandaigua and Rochester) in western New York State were sampled as part of an ongoing EPRI background study of PAHs in surface soil. Each municipality sampled has a population of at least 10,000 people and a population density of at least 1000 people per square mile. At each site, two locations were sampled from 0 – 2.54 cm below grade and from 2.54 – 15 cm below grade, after removing surface vegetation, if present. At all sites, samples were composited to generate one sample representative of 0 – 15 cm below grade.
The patterns of PAHs seen in the samples were consistent with those in surface soil samples containing PAHs from pyrogenic sources that have been subjected to environmental weathering. To compare the concentrations of individual PAHs for all of the sites, summary statistics were performed, with not-detected results equal to one half of the detection limit. The results for the individual PAHs were generally lognormally distributed. Benzo(a)pyrene concentrations ranged from a non-detectable 7.5 – 4740 µg/kg, with a median concentration of 431 µg/kg. The upper quartile concentration was 1040 µg/kg, while the upper 95th percentile concentration was 2770 µg/kg. These concentrations appear to be consistent with literature-reported levels for anthropogenic background in small- to medium-sized residential, commercial, and light industrial areas, but notably lower than background observed in more highly populated cities and commercial/industrial areas.
ix
CONTENTS
1 INTRODUCTION ....................................................................................................................1-1 Background ...........................................................................................................................1-1 Objectives .............................................................................................................................1-2 Summary of Approach...........................................................................................................1-2 EPRI Perspective ..................................................................................................................1-3 Report Organization ..............................................................................................................1-3
2 SAMPLING AND ANALYSIS METHODS ..............................................................................2-1 Definitions .............................................................................................................................2-1 Sampling Design ...................................................................................................................2-1
Selecting Areas to Sample ...............................................................................................2-2 Selecting Sites to Sample.................................................................................................2-2 Selecting Locations to Sample at Each Site.....................................................................2-2
Sample Compositing and Analysis Scheme..........................................................................2-2 Composites.......................................................................................................................2-2 Sample Analysis ...............................................................................................................2-3 Data Management and Analysis.......................................................................................2-3
3 PRESENTATION OF RESULTS ............................................................................................3-1 Field Activities .......................................................................................................................3-1
Site and Area Land Use Types.........................................................................................3-1 Summary of Physical Sample Characteristics.......................................................................3-3 Sample Results for PAHs......................................................................................................3-3
Summary Statistics for PAH Results from All Sites ..........................................................3-4 Quality Assurance/Quality Control (QA/QA) Results.............................................................3-7
4 CONCLUSIONS .....................................................................................................................4-1
5 REFERENCES .......................................................................................................................5-1
A TABLES ................................................................................................................................ A-1
xi
LIST OF FIGURES
Figure 3-1 Example of Municipal Site – Department of Public Works .......................................3-2 Figure 3-2 Example of a Heavy Residential Area Use Classification – Right of Way Site.........3-3 Figure 3-3 Normal Histogram and Probability Plot for Pyrene ...................................................3-8 Figure 3-4 Normal Histogram and Probability Plot for Benzo(a)pyrene.....................................3-9 Figure 3-5 Lognormal Histogram and Probability Plot for Pyrene............................................3-10 Figure 3-6 Lognormal Histogram and Probability Plot for Benzo(a)pyrene..............................3-11
xiii
LIST OF TABLES
Table 3-1 Summary Statistics for Raw PAH Data in Western New York (µg/kg).......................3-5 Table 3-2 Summary Statistics of Normalized PAH Data in Western New York (µg/kg)...........3-12 Table 3-3 Back-Transformed PAH Statistics in Western New York (µg/kg) ............................3-14 Table 4-1 Summary Benzo(a)pyrene Statistics for Other Data Sets Compared to this
Study (µg/kg)......................................................................................................................4-2 Table A-1 Sample Soil Descriptions, Total Organic Carbon, and Percent Solids.................... A-1 Table A-2 Key to Abbreviations for PAHs ................................................................................ A-2 Table A-3 Polycyclic Aromatic Hydrocarbon (PAH) Results in Background Surface Soils ...... A-3 Table A-4 Field Duplicate Results with Relative Percent Differences (RPDs)......................... A-7 Table A-5 Laboratory Duplicate Results with Relative Percent Differences (RPDs)................ A-8
1-1
1 INTRODUCTION
This report provides an update of ongoing EPRI research that is being performed to examine the distribution of polycyclic aromatic hydrocarbon (PAH) concentrations in surface soils of populated areas. For inclusion in this study, all of the areas sampled had populations of at least 10,000 people and population densities of at least 1,000 people per square mile. This report summarizes the purposes of this research, the procedures used during the study, as well as the results and conclusions obtained from recently collected data in western New York State. Results from the first 18 municipalities in Illinois (16) and central New York (2) sampled for this study are presented in the report entitled Polycyclic Aromatic Hydrocarbons (PAHs) in Surface Soil (EPRI, 2002).
Background
The natural and anthropogenic sources of PAHs to the environment are many and diverse. Consequently, PAHs are found in most samples of soil, water, and biota throughout the world (Blumer, 1976). The frequent detection of PAHs in surface soil and sediment is of particular importance for environmental investigations and cleanups, because the concentrations of PAHs often define the extent of contamination and influence the estimated risk from contamination at a variety of sites.
Further, at many sites, it becomes very important to be able to determine where the site-specific contamination ends and where the local background PAHs begin. Although there are many literature references concerning the presence of PAHs in environmental media, including surface soils, a review of those references indicated that there is a lack of consistent data on distributions and concentrations of PAHs in surface soils (EPRI, 2000). This study was designed to generate an internally consistent, rugged set of PAH concentrations in surface soils from a statistically significant number of locations in populated areas above a set size and density. The samples were collected and analyzed so that PAH data collected at different times and from different locations would be comparable, and would not contain unacceptable sampling or analytical bias.
Introduction
1-2
Objectives
The overall objectives of the EPRI study of background PAHs were as follows:
1. Collect, composite (if appropriate), analyze, and interpret the results of surface soil samples from commercial/industrial, residential, municipal, and some rural locations.
2. Compile the data and determine their distribution and statistics. Correlate the PAH concentrations and patterns to site conditions or histories, when possible. For example, determine if PAH distributions and concentrations observed in soils near a municipal parking lot in one urban area are similar to PAH distributions and concentrations found in soils near a municipal parking lot in another urban area in a different part of the country.
3. Generate detailed information of typical PAH levels found at different locations as an aid in establishing appropriate and realistic cleanup levels for surface soil PAHs at sites undergoing remedial actions.
Summary of Approach
The following list gives general descriptions of the sampling and analysis approach.
• Thus far, samples have been collected from sites in the States of New York and Illinois. The sites were selected from urban areas and included utility-owned sites, parks, forested areas, roadside rights of way, fields, public properties, and selected commercial properties.
• For each state or region, a pseudo-random process was used to select the sample locations. The process involved randomly selecting populated areas within each state, randomly selecting a pre-determined number of sites within each area, and then selecting sampling locations at each site that were representative of the site and not impacted by known sources of PAHs.
• Sample collection locations were selected to be representative of the overall site conditions based on a visual assessment of the site and field judgment. Samples were not collected in immediate proximity to known sources of PAHs, such as railroad tracks.
• Since sites varied significantly in size, shape, and proximity to potential sources of PAHs, the sampling configurations were site-specific. The sample configurations were designed so that the soil samples collected were representative of the site as a whole or a specific portion of the site, if appropriate. Sampling locations for each site were kept in relatively close proximity to each other so that compositing these samples would be appropriate.
• Selected samples were composited in the laboratory prior to analysis, with only two and rarely three samples combined into one composite sample. Samples were analyzed for PAHs and total organic carbon (TOC).
• The precision and accuracy of the data were estimated using the results of quality control samples. The distribution and basic statistics of the data were calculated and reported. Also, the data were carefully examined for factors that correlate with the PAH concentrations, such as land use types and proximity to identifiable PAH sources.
Introduction
1-3
EPRI Perspective
Natural and anthropogenic sources of polycyclic aromatic hydrocarbons (PAHs) in the environment are quite diverse due to many different factors of human activity or inactivity in an area. Consequently, PAHs are found in many samples of soil, water, and biota throughout the world. The frequent detection of PAHs in surface soil and sediment samples is of particular importance for environmental investigations and cleanups, because the concentrations of PAHs often define the extent of contamination and influence the estimated risk from contamination at a variety of sites. At many sites, it becomes very important to be able to determine where the site-specific contamination ends and where the local background PAHs begin. The EPRI Background PAH Study was designed to generate an internally consistent; rugged set of PAH concentrations in surface soils from numerous locations covering a range of land uses. This specific project completes the data set of a former effort to sample background soils in western and northwestern New York State. (EPRI Tech Report 1005295).
Report Organization
Section 2 presents the sampling and analysis methods used for this study, including the sampling design. The results obtained from this work are presented in Section 3. Section 4 contains a discussion of conclusions based on the results from the study. The references used in the development of this document are shown in Section 5.
2-1
2 SAMPLING AND ANALYSIS METHODS
Definitions
• Sample Location - the actual place where an individual sample was collected. Each sample location corresponds with the sample identification number on its sample jar and on the chain-of-custody form.
• Site - the property or plot of land containing one or more sample locations. For example, “Chestnut Hill Park” or “highway 1 median” was designated as the “site”.
• Area - A city, town, county, or other locality that may contain multiple sites.
• Urbanized Area (UA) - an area consisting of a central place and adjacent urban fringe that together have a minimum residential population of at least 50,000 people and an overall population density of 1,000 people per square mile of land area (U.S. Census Bureau, 2000).
• Populated Area - an area that has a population density of 1,000 people per square mile of land area (equivalent to the UA), but with a minimum residential population of 10,000 people. The populated area designation includes smaller cities and towns that have significant urban centers, but do not meet the U.S. Census Bureau’s definition of an urbanized area.
Sampling Design
A layered sampling design based on random sampling was used for the study. For western New York, two cities were selected for sampling based in part on their position on a randomly ranked list of populated areas in the service area of the participating utility, as well as judgments about the ease of gaining permission from city officials to perform the sampling and the value of performing a study in that area for the utility.
Next, a similar pseudo-random selection scheme was used to select a pre-determined number of sites within each area. The number of sites was based on an estimate of the number of samples needed for statistical significance and the number of areas chosen for the particular state or region.
As was generally the case during previous sampling, samples were collected from two areal locations at each site. The field engineer chose the sample locations at each site at the time of sample collection.
Sampling and Analysis Methods
2-2
Selecting Areas to Sample
In New York, two populated areas were selected for sampling based on a combination of factors. First, only urbanized and populated areas within the service area of the participating utility were included in the mix of possible sampling areas. Then, the random number generator function in Microsoft® Excel was used to rank these areas. Finally, the principal investigators made judgments with the participating utility to determine the practicability of each potential sampling area. Based on these judgments, two areas were selected for sampling.
Selecting Sites to Sample
Once areas were selected for sampling, road maps and USGS 7.5 minute quad(s) for each area were used to select the actual sites to be sampled. The sites within each area were also selected randomly. A coordinate system was laid out over maps of each area and a random number generator was used to select points that fell on this coordinate system. In the field, the closest suitable site to each selected point was sampled. Suitable sampling sites included parks, roadway medians, utility rights-of-way, municipal properties, residential properties, parking lot buffers, or vacant lots. In western New York, 20 sites were sampled in one urbanized area (Rochester), while 10 sites were sampled in one populated area (Canandaigua).
Selecting Locations to Sample at Each Site
Two locations were sampled at each site and were selected by the field engineer to be representative of the overall site conditions based on a visual assessment of the site and judgment. The field engineer considered the area of the site, obstructions, visible evidence of contamination, and other practical matters. Samples were not collected in immediate proximity to known sources of PAHs, such as railroad tracks.
Sample Compositing and Analysis Scheme
Composites
At every location in this study, samples were collected from both 0 to 2.54 cm below grade and 2.54 to 15 cm below grade. The samples were labeled and shipped to META Environmental, Inc. in Watertown, Massachusetts for analysis. As described in the QAPP, all of the field samples were composited in the laboratory to decrease the number of analyses.
The samples from all 30 sites in western New York were composited in the same way. At each site, the four discrete samples collected were composited to create one sample that represented 0 to 15 cm for that site. Specifically, 10 grams from each of the 0 to 2.54 cm samples was mixed with 50 grams from each of the 2.54 to 15 cm samples to create a composite sample representative of the interval from 0 to 15 cm.
The original discrete samples are stored frozen for possible future analysis.
Sampling and Analysis Methods
2-3
Sample Analysis
Each composited soil sample was analyzed for PAHs using vigorous Soxhlet extraction, EPA Method 3540C. A portion of each extract was cleaned up using silica gel columns (EPA Method 3630C mod.). The cleaned extracts were analyzed using GC/MS, EPA Method 8270 for PAHs only. (U.S. EPA, 1986)
Lastly, samples were analyzed for total organic carbon (TOC) using a modification (digestion with additional heat) of the Walkley-Black method (90-3, 1963).
Data Management and Analysis
All data were checked for errors and quality control acceptability. Deviations from the data quality objectives were noted on raw data worksheets. No quality control deviations were identified that were significant enough to reject any data. The sample identifiers, sample descriptions, and final data were entered into a database constructed using Microsoft® Access software. The data were coded so that simple queries could be used to extract subsets of the data, such as by depth or by populated area. For the purposes of this report, the data were exported to Microsoft® Excel spreadsheets for formatting.
All statistical and graphical analyses were conducted using commercial software (Statistica, version 6.0, from StatSoft, Inc.)
3-1
3 PRESENTATION OF RESULTS
Field Activities
For western New York, samples were collected from the cities of Canandaigua and Rochester. On July 29, 2003, ten sites in Canandaigua were sampled. From July 30 to August 1, 2003, 20 sites were sampled in Rochester. Two locations were sampled from each of the 30 sites samples in these two cities. Random locations were generated on maps of these two cities, and the nearest appropriate sampling locations were identified in the field.
The sampling procedures used for all of the locations in this study were the same. At each location, samples were collected with a clean stainless steel trowel from 0 to 2.54 cm and from 2.54 to 15 cm below grade, after peeling back surface vegetation, if present. Next, the soil from each interval was placed in a stainless steel bowl and homogenized. Then, the homogenized soil was placed in 4 oz soil jars with Teflon-lined screw caps. Lastly, the jars were labeled and placed on ice for shipment to the laboratory.
Site and Area Land Use Types
A total of 30 sites were sampled in western New York, including a variety of different types of sites. Based on observations made in the field, a subjective classification of the land use was made for each site sampled as part of this study. The site use classifications included recreational (40%), rights of way (30%), municipal (27%), and conservation (3%). No commercial, industrial, open land, residential, or utility sites were sampled in western New York. Recreational sites included ball fields and recreational parks. Sites where a municipal right of way existed, such as the areas near streets, were designated as rights of way even if they formed part of a residential or commercial lawn. The municipal designation was used for public areas such as police stations, fire stations, and town buildings, including schools. An example of a municipal site (Department of Public Works) is shown in Figure 3-1. Conservation sites included areas such as forest and nature preserves. Sites were considered utility sites either if they were owned by a utility or if there was a utility right of way in the area sampled. The residential sites were home lots or apartment complexes. Sites with businesses on them were designated as commercial for retail business, or as industrial for manufacturing or similar operations. The open land designation was used for land with no specific purpose or structures.
In addition to site use classifications, a subjective characterization of the surrounding area use was made at each of the 30 sites. The resulting area use classifications included heavy residential (57%), commercial (23%), light industrial (13%), heavy industrial (3%), and rural (3%). No sites were sampled in agricultural or light residential areas in western New York.
Presentation of Results
3-2
Areas were considered heavy residential if houses were present at a density of at least one house per acre. Figure 3-2 shows an example of a site in an area designated as heavy residential. Otherwise, if the houses were sparse, the area was assigned the light residential designation. Similarly, areas with a high density of non-retail businesses were considered heavy industrial, particularly if manufacturing was present. On the other hand, areas with just one or a few non- retail businesses were considered light industrial. Commercial was used to describe areas with retail businesses. For sites bordering farmland, the area was considered agricultural. Lastly, areas with very few businesses or residences, and without farmland, were designated as rural.
Figure 3-1 Example of Municipal Site – Department of Public Works
Presentation of Results
3-3
Figure 3-2 Example of a Heavy Residential Area Use Classification – Right of Way Site
Summary of Physical Sample Characteristics
Table A-1 (Appendix A) lists the sample identification, soil description, total organic carbon (TOC) content, and percent solids content for each of the samples analyzed. Consistent with the intent of the sampling plan, the samples contained various amounts of silt, sand, clay and gravel. For the 30 samples analyzed from western New York, the TOC levels ranged from 0.6% to slightly more than 10% with a median TOC of 3.1%. Similarly, the TOC levels from previously reported samples (EPRI, 2002) ranged from 0.5% to slightly more than 11% with a median TOC of 2.4%. The percent solids content was generally 80 to 95%, with a median of 90%, which was slightly higher overall compared with the previous data set (EPRI, 2002) but not unexpectedly, given the dry soil conditions at the time of the western New York sampling.
Sample Results for PAHs
The names of the PAHs have been codified in some tables and figures for ease of formatting. Table A-2 (Appendix A) shows the PAH names and codes used in this report.
Table A-3 (Appendix A) presents the full data set of PAH concentrations for the 30 sites sampled. Non-detects are reported as not detected below the sample-specific quantitation limits, as indicated by the “U” flag next to the concentration.
Presentation of Results
3-4
Summary Statistics for PAH Results from All Sites
For the previously reported data (EPRI, 2002), a reduced data set was generated for most of the statistical and graphical analyses that were conducted. The reduced data set was generated in two ways. First, one half the sample-specific quantitation limit was substituted for non-detects. Then, arithmetic composites were generated where the 0 to 2.54 cm and 2.54 to 15 cm samples were analyzed separately. This was done by generating weighted averages based on the sampling intervals, which are representative of the 0 to 15 cm interval at those locations.
For western New York, all of the samples analyzed were composite samples from 0 to 15 cm below grade. Table 3-1 lists the summary description statistics for all 30 samples. Included in the statistics for each compound are the mean, 95% confidence intervals of the mean, geometric mean, median, minimum and maximum values, range, upper and lower quartiles, quartile range, 5th and 95th percentiles, variance, standard deviation, standard error, skewness, and kurtosis. All statistics reported in Table 3-1 were performed on the raw data, assuming that the concentrations of each variable (PAH analyte) are normally distributed. As shown in Table 3-1, benzo(a)pyrene concentrations ranged from 7.5 to 4,740 µg/kg with a median concentration of 431 µg/kg. The upper quartile concentration was 1,040 µg/kg, while the upper 95th percentile concentration was 2,770 µg/kg. These concentrations appear to be consistent with literature-reported levels for anthropogenic background in small to medium-sized residential, commercial, and light industrial areas, but notably lower than background observed in more highly populated cities and commercial/industrial areas (Bradley et al., 1994 and MA DEP, 2002).
Pre
sent
atio
n of
Res
ults
3-5
Tabl
e 3-
1 S
umm
ary
Sta
tistic
s fo
r Raw
PA
H D
ata
in W
este
rn N
ew Y
ork
(µg/
kg)
PA
Hs
Val
id N
M
ean
Low
er 9
5%
Con
fiden
ce
Upp
er 9
5%
Con
fiden
ce
Geo
met
ric
Mea
n M
edia
n M
inim
um
Max
imum
R
ange
NA
P
30
100
47.1
15
3 51
.8
42.6
8.
01
667
659
2MN
30
89
.8
27.2
15
2 30
.3
29.8
2.
25
816
814
AC
Y
30
119
55.1
18
3 42
.6
52.8
3.
18
713
710
AC
E
30
47.1
20
.9
73.3
19
.2
15.5
2.
67
260
257
FLU
30
55
.3
26
84.7
23
.5
24.7
1.
91
319
317
PH
E
30
797
400
1,19
0 32
8 35
3 6.
48
4,82
0 4,
810
AN
T 30
20
5 99
.7
311
78.8
83
.8
1.37
1,
330
1,33
0
FLR
30
1,
470
765
2,17
0 67
7 71
5 19
.5
9,01
0 8,
990
PY
R
30
1,16
0 61
5 1,
700
542
553
17
6,75
0 6,
730
BA
A
30
690
363
1,02
0 29
8 33
0 8.
44
3,91
0 3,
900
CH
R
30
794
434
1,15
0 37
6 39
1 11
.6
4,33
0 4,
320
BB
F 30
72
8 40
5 1,
050
308
358
5.3
3,34
0 3,
330
BK
F 30
61
3 34
4 88
1 30
6 38
3 9.
36
3,35
0 3,
340
BA
P
30
830
440
1,22
0 37
0 43
1 7.
48
4,74
0 4,
730
IND
30
50
2 25
3 75
2 19
3 24
6 5.
03
2,33
0 2,
320
DA
A
30
181
92
270
68.6
88
.8
1.83
81
9 81
7
BG
P
30
541
280
801
223
247
7.28
2,
420
2,41
0
All
stat
istic
s ca
lcul
ated
with
not
-det
ecte
d re
sults
equ
al to
one
hal
f of
the
dete
ctio
n lim
its, w
hich
are
sho
wn
for
each
sam
ple
in T
able
A-3
.
P
rese
ntat
ion
of R
esul
ts
3-6
Tabl
e 3-
1 (c
ont.)
S
umm
ary
Sta
tistic
s fo
r Raw
PA
H D
ata
in W
este
rn N
ew Y
ork
(µg/
kg)
PA
Hs
Low
er
Qua
rtile
U
pper
Q
uart
ile
Qua
rtile
R
ange
5t
h P
erce
ntile
95
th
Per
cent
ile
Var
ianc
e S
tand
ard
Dev
iatio
n S
tand
ard
Err
or
Ske
wne
ss
Kur
tosi
s
NA
P
21.5
11
4 92
.5
10.3
44
3 20
,200
14
2 26
2.
86
9
2MN
10
.3
68
57.7
2.
51
360
28,1
00
168
30.6
3.
29
12.2
AC
Y
8.36
12
8 12
0 4.
06
514
29,2
00
171
31.2
2.
16
4.64
AC
E
7.2
51.8
44
.6
3.78
23
1 4,
920
70.1
12
.8
2.09
3.
6
FLU
8.
02
79.4
71
.4
3.93
25
4 6,
180
78.6
14
.4
2.18
4.
49
PH
E
107
1,07
0 96
3 14
.9
2,82
0 1,
130,
000
1,06
0 19
4 2.
33
6.39
AN
T 20
.3
317
297
10.3
59
2 80
,100
28
3 51
.7
2.47
7.
77
FLR
24
0 1,
760
1,52
0 27
.6
4,97
0 3,
540,
000
1,88
0 34
4 2.
58
8.32
PY
R
199
1,45
0 1,
250
24.1
4,
090
2,12
0,00
0 1,
460
266
2.38
6.
84
BA
A
95.2
98
3 88
8 9.
01
2,25
0 76
6,00
0 87
5 16
0 2.
16
5.43
CH
R
129
1,12
0 99
1 18
.3
2,49
0 93
0,00
0 96
4 17
6 2.
1 5.
23
BB
F 11
1 96
8 85
7 10
.3
2,37
0 74
8,00
0 86
5 15
8 1.
55
1.73
BK
F 10
6 74
9 64
3 13
.7
1,76
0 51
9,00
0 72
0 13
1 2.
24
6.25
BA
P
135
1,04
0 90
5 10
.6
2,77
0 1,
090,
000
1,04
0 19
1 2.
25
5.95
IND
81
.3
489
408
8.17
2,
170
446,
000
667
122
1.76
2.
02
DA
A
27.3
18
7 16
0 2.
3 81
8 56
,900
23
8 43
.5
1.75
2.
09
BG
P
95.8
54
2 44
6 12
.9
2,10
0 48
7,00
0 69
8 12
7 1.
65
1.57
All
stat
istic
s ca
lcul
ated
with
not
-det
ecte
d re
sults
equ
al to
one
hal
f of
the
dete
ctio
n lim
its, w
hich
are
sho
wn
for
each
sam
ple
in T
able
A-3
.
Presentation of Results
3-7
The differences between the mean concentrations and the median concentrations, as well as the skewness value (skewness near zero is indicative of a normal distribution) strongly suggest that the data are not normally distributed. This observation was explored further using normal probability plots and frequency histograms. Figures 3-4 and 3-5 show normal probability plots and frequency histograms for pyrene and benzo(a)pyrene that clearly indicate the lognormal distribution of the data for those analytes. Similar plots were developed for the other analytes with similar results, but are not included in this report. Considering the underlying distributions of the data, the summary statistics for the full data set were recalculated on the natural logarithms of the raw data and then back-transformed to concentrations.
Figures 3-6 and 3-7 show the normal probability plots and frequency histograms for the log-transformed concentration data of pyrene and benzo(a)pyrene. In contrast to the raw concentration data, the log-transformed data are clearly normally distributed.
Table 3-2 shows the summary description statistics for the data set determined from the log-normalized data and Table 3-3 shows these values back-transformed to concentrations. The values for skewness and kurtosis are generally near zero, which suggests that the concentrations of the various PAHs in the samples have a lognormal distribution. As shown in Table 3-3, statistics such as the median, quartiles, and percentiles do not change when using transformed data because they are not based on parametric statistics. However, the mean, variance, standard error, and confidence intervals about the mean change substantially when using log-transformed data. Again, these concentrations appear to be consistent with literature-reported levels for anthropogenic background in small to medium-sized residential, commercial, and light industrial areas, but notably lower than background observed in more highly populated cities and commercial/industrial areas.
As discussed above, the individual PAH distributions are generally lognormal. In particular, analytes for which there were few non-detects, such as pyrene, clearly show lognormal distributions. However, some deviations from a lognormal distribution were noted. Specifically, analytes for which there were relatively many non-detects and low concentrations, such as acenaphthene and naphthalene, are not as well behaved. This is common for data sets with many measurements near or below the detection limits.
Quality Assurance/Quality Control (QA/QA) Results
Laboratory QC followed the Quality Assurance Project Plan (QAPP) with regard to holding times, blanks, surrogate and matrix spikes, and replicates. The results of laboratory QC checks were well within the QAPP-specified criteria with few exceptions.
The reporting limits were based on the sample equivalent of the lowest linear calibration standard. The reporting limits ranged from about 4 µg/kg to about 20 µg/kg, depending on the compound and sample conditions. Detection limits were about one half of the reporting limits and based on actual signal to noise ratios.
Presentation of Results
3-8
Normal P-Plot: Pyrene
-1000 0 1000 2000 3000 4000 5000 6000 7000 8000
Concentration (ug/kg)
-3
-2
-1
0
1
2
3
4
Expe
cted
Nor
mal
Val
ue
Histogram: Pyrene Expected Normal
-1000 0 1000 2000 3000 4000 5000 6000 7000
X <= Category Boundary (Concentration ug/kg)
0
2
4
6
8
10
12
14
16
18
20
22
Num
ber o
f Obs
erva
tions
Figure 3-3 Normal Histogram and Probability Plot for Pyrene
Presentation of Results
3-9
Normal P-Plot: Benzo(a)pyrene
-1000 0 1000 2000 3000 4000 5000
Concentration (ug/kg)
-3
-2
-1
0
1
2
3
4
Expe
cted
Nor
mal
Val
ue
Histogram: Benzo(a)pyrene Expected Normal
-500 0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000
X <= Category Boundary (Concentration ug/kg)
0
2
4
6
8
10
12
14
16
18
Num
ber o
f Obs
erva
tions
Figure 3-4 Normal Histogram and Probability Plot for Benzo(a)pyrene
Presentation of Results
3-10
Normal P-Plot: Pyrene
2 3 4 5 6 7 8 9 10
Normalized Concentration (ug/kg)
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
Expe
cted
Nor
mal
Val
ue
Histogram: Pyrene Expected Normal
2 3 4 5 6 7 8 9
X <= Category Boundary (Normalized concentration ug/kg)
0
1
2
3
4
5
6
7
8
9
10N
umbe
r of O
bser
vatio
ns
Figure 3-5 Lognormal Histogram and Probability Plot for Pyrene
Presentation of Results
3-11
Normal P-Plot: Benzo(a)pyrene
1 2 3 4 5 6 7 8 9
Normalized Concentration (ug/kg)
-3.0
-2.5
-2.0
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
Expe
cted
Nor
mal
Val
ue
Histogram: Benzo(a)pyrene Expected Normal
1 2 3 4 5 6 7 8 9
X <= Category Boundary (Normalized concentration ug/kg)
0
1
2
3
4
5
6
7
8
9
10
11N
umbe
r of O
bser
vatio
ns
Figure 3-6 Lognormal Histogram and Probability Plot for Benzo(a)pyrene
P
rese
ntat
ion
of R
esul
ts
3-12
Tabl
e 3-
2 S
umm
ary
Sta
tistic
s of
Nor
mal
ized
PA
H D
ata
in W
este
rn N
ew Y
ork
(µg/
kg)
PA
Hs
Val
id N
M
ean
Low
er 9
5%
Con
fiden
ce
Upp
er 9
5%
Con
fiden
ce
Geo
met
ric
Mea
n M
edia
n M
inim
um
Max
imum
R
ange
NA
P
30
3.95
3.
52
4.37
3.
79
3.75
2.
08
6.5
4.42
2MN
30
3.
41
2.86
3.
96
3.07
3.
36
0.81
6.
7 5.
89
AC
Y
30
3.75
3.
16
4.34
3.
38
3.92
1.
16
6.57
5.
41
AC
E
30
2.95
2.
45
3.45
2.
66
2.74
0.
98
5.56
4.
58
FLU
30
3.
16
2.65
3.
66
2.84
3.
21
0.65
5.
77
5.12
PH
E
30
5.79
5.
22
6.37
5.
55
5.86
1.
87
8.48
6.
61
AN
T 30
4.
37
3.77
4.
96
3.92
4.
43
0.31
7.
19
6.88
FLR
30
6.
52
5.98
7.
06
6.34
6.
57
2.97
9.
11
6.14
PY
R
30
6.3
5.77
6.
83
6.11
6.
32
2.83
8.
82
5.98
BA
A
30
5.7
5.13
6.
26
5.45
5.
8 2.
13
8.27
6.
14
CH
R
30
5.93
5.
4 6.
45
5.74
5.
97
2.45
8.
37
5.92
BB
F 30
5.
73
5.14
6.
32
5.46
5.
87
1.67
8.
11
6.45
BK
F 30
5.
72
5.21
6.
24
5.53
5.
94
2.24
8.
12
5.88
BA
P
30
5.91
5.
35
6.48
5.
67
6.06
2.
01
8.46
6.
45
IND
30
5.
26
4.66
5.
86
4.96
5.
5 1.
62
7.75
6.
14
DA
A
30
4.23
3.
62
4.84
3.
77
4.49
0.
6 6.
71
6.1
BG
P
30
5.41
4.
85
5.97
5.
17
5.51
1.
99
7.79
5.
81
All
stat
istic
s ca
lcul
ated
with
not
-det
ecte
d re
sults
equ
al to
one
hal
f of
the
dete
ctio
n lim
its, w
hich
are
sho
wn
for
each
sam
ple
in T
able
A-3
.
Pre
sent
atio
n of
Res
ults
3-13
Tabl
e 3-
2 (c
ont.)
S
umm
ary
Sta
tistic
s of
Nor
mal
ized
PA
H D
ata
in W
este
rn N
ew Y
ork
(µg/
kg)
PA
Hs
Low
er
Qua
rtile
U
pper
Q
uart
ile
Qua
rtile
R
ange
5t
h P
erce
ntile
95
th
Per
cent
ile
Var
ianc
e S
tand
ard
Dev
iatio
n S
tand
ard
Err
or
Ske
wne
ss
Kur
tosi
s
NA
P
3.07
4.
74
1.67
2.
33
6.09
1.
29
1.14
0.
21
0.39
-0
.45
2MN
2.
33
4.22
1.
89
0.92
5.
89
2.16
1.
47
0.27
0.
38
-0.2
7
AC
Y
2.12
4.
85
2.73
1.
4 6.
24
2.51
1.
58
0.29
0.
01
-1.1
5
AC
E
1.97
3.
95
1.97
1.
33
5.44
1.
79
1.34
0.
24
0.55
-0
.82
FLU
2.
08
4.37
2.
29
1.37
5.
54
1.85
1.
36
0.25
0.
23
-0.8
3
PH
E
4.67
6.
98
2.3
2.7
7.94
2.
38
1.54
0.
28
-0.5
0.
19
AN
T 3.
01
5.76
2.
75
2.33
6.
38
2.52
1.
59
0.29
-0
.36
-0.1
6
FLR
5.
48
7.47
1.
99
3.32
8.
51
2.07
1.
44
0.26
-0
.6
0.35
PY
R
5.29
7.
28
1.99
3.
18
8.32
2.
02
1.42
0.
26
-0.5
7 0.
26
BA
A
4.56
6.
89
2.33
2.
2 7.
72
2.3
1.52
0.
28
-0.5
8 0.
18
CH
R
4.86
7.
02
2.16
2.
91
7.82
1.
97
1.4
0.26
-0
.53
0.23
BB
F 4.
71
6.88
2.
17
2.33
7.
77
2.49
1.
58
0.29
-0
.69
0.38
BK
F 4.
66
6.62
1.
96
2.62
7.
47
1.89
1.
37
0.25
-0
.69
0.54
BA
P
4.91
6.
95
2.04
2.
36
7.93
2.
27
1.51
0.
28
-0.7
5 0.
79
IND
4.
4 6.
19
1.79
2.
1 7.
68
2.59
1.
61
0.29
-0
.44
-0.2
1
DA
A
3.31
5.
23
1.92
0.
83
6.71
2.
69
1.64
0.
3 -0
.48
-0.2
2
BG
P
4.56
6.
3 1.
73
2.56
7.
65
2.25
1.
5 0.
27
-0.3
-0
.32
All
stat
istic
s ca
lcul
ated
with
not
-det
ecte
d re
sults
equ
al to
one
hal
f of
the
dete
ctio
n lim
its, w
hich
are
sho
wn
for
each
sam
ple
in T
able
A-3
.
P
rese
ntat
ion
of R
esul
ts
3-14
Tabl
e 3-
3 B
ack-
Tran
sfor
med
PA
H S
tatis
tics
in W
este
rn N
ew Y
ork
(µg/
kg)
PA
Hs
Val
id N
M
ean
Low
er 9
5%
Con
fiden
ce
Upp
er 9
5%
Con
fiden
ce
Geo
met
ric
Mea
n M
edia
n M
inim
um
Max
imum
R
ange
NA
P
30
51.8
33
.9
79.2
44
.2
42.6
8.
01
667
83.3
2MN
30
30
.3
17.5
52
.5
21.5
29
.8
2.25
81
6 36
3
AC
Y
30
42.6
23
.6
77
29.5
52
.8
3.18
71
3 22
4
AC
E
30
19.2
11
.6
31.6
14
.4
15.5
2.
67
260
97.4
FLU
30
23
.5
14.1
39
17
.1
24.7
1.
91
319
167
PH
E
30
328
184
583
257
353
6.48
4,
820
744
AN
T 30
78
.8
43.5
14
3 50
.5
83.8
1.
37
1,33
0 97
1
FLR
30
67
7 39
6 1,
160
565
715
19.5
9,
010
462
PY
R
30
542
319
922
452
553
17
6,75
0 39
7
BA
A
30
298
169
525
234
330
8.44
3,
910
463
CH
R
30
376
222
634
310
391
11.6
4,
330
373
BB
F 30
30
8 17
1 55
5 23
4 35
8 5.
3 3,
340
630
BK
F 30
30
6 18
3 51
1 25
1 38
3 9.
36
3,35
0 35
8
BA
P
30
370
211
650
291
431
7.48
4,
740
634
IND
30
19
3 10
6 35
2 14
3 24
6 5.
03
2,33
0 46
3
DA
A
30
68.6
37
.2
127
43.2
88
.8
1.83
81
9 44
8
BG
P
30
223
128
391
177
247
7.28
2,
420
332
All
stat
istic
s ca
lcul
ated
with
not
-det
ecte
d re
sults
equ
al to
one
hal
f of
the
dete
ctio
n lim
its, w
hich
are
sho
wn
for
each
sam
ple
in T
able
A-3
.
Pre
sent
atio
n of
Res
ults
3-15
Tabl
e 3-
3 (c
ont.)
B
ack-
Tran
sfor
med
PA
H S
tatis
tics
in W
este
rn N
ew Y
ork
(µg/
kg)
PA
Hs
Low
er
Qua
rtile
U
pper
Q
uart
ile
Qua
rtile
R
ange
5t
h P
erce
ntile
95
th
Per
cent
ile
Var
ianc
e S
tand
ard
Dev
iatio
n S
tand
ard
Err
or
Ske
wne
ss
Kur
tosi
s
NA
P
21.5
11
4 5.
3 10
.3
443
3.63
3.
11
1.23
1.
48
0.64
2MN
10
.3
68
6.6
2.51
36
0 8.
69
4.35
1.
31
1.46
0.
76
AC
Y
8.36
12
8 15
.3
4.06
51
4 12
.3
4.88
1.
34
1.01
0.
32
AC
E
7.2
51.8
7.
19
3.78
23
1 5.
97
3.81
1.
28
1.73
0.
44
FLU
8.
02
79.4
9.
9 3.
93
254
6.38
3.
9 1.
28
1.26
0.
44
PH
E
107
1,07
0 10
14
.9
2,82
0 10
.8
4.67
1.
33
0.61
1.
2
AN
T 20
.3
317
15.6
10
.3
592
12.5
4.
9 1.
34
0.7
0.85
FLR
24
0 1,
760
7.33
27
.6
4,97
0 7.
96
4.22
1.
3 0.
55
1.41
PY
R
199
1,45
0 7.
29
24.1
4,
090
7.52
4.
14
1.3
0.56
1.
3
BA
A
95.2
98
3 10
.3
9.01
2,
250
9.99
4.
56
1.32
0.
56
1.2
CH
R
129
1,12
0 8.
68
18.3
2,
490
7.17
4.
07
1.29
0.
59
1.25
BB
F 11
1 96
8 8.
72
10.3
2,
370
12
4.84
1.
33
0.5
1.46
BK
F 10
6 74
9 7.
07
13.7
1,
760
6.6
3.95
1.
29
0.5
1.72
BA
P
135
1,04
0 7.
7 10
.6
2,77
0 9.
7 4.
52
1.32
0.
47
2.21
IND
81
.3
489
6.01
8.
17
2,17
0 13
.4
5.01
1.
34
0.64
0.
81
DA
A
27.3
18
7 6.
85
2.3
818
14.7
5.
15
1.35
0.
62
0.8
BG
P
95.8
54
2 5.
66
12.9
2,
100
9.52
4.
49
1.32
0.
74
0.72
All
stat
istic
s ca
lcul
ated
with
not
-det
ecte
d re
sults
equ
al to
one
hal
f of
the
dete
ctio
n lim
its, w
hich
are
sho
wn
for
each
sam
ple
in T
able
A-3
.
Presentation of Results
3-16
Two field duplicate samples were collected in western New York. Field duplicates were samples from the same site and interval placed in separate jars in the field by the field engineer. They were shipped to the laboratory separately and processed by the laboratory as independent samples. The precision of soil duplicates was measured using the relative percent difference (RPD) statistic. This statistic is the range divided by the mean times 100. The RPDs varied by sample and by compound, but were all less than 50%. Table A-4 shows the field duplicate results.
Two laboratory duplicate samples were prepared in the laboratory by analyzing two separate aliquots of two composited samples from western New York. The laboratory duplicates provide information on the combined effects of the effectiveness of the soil homogenization step and the variability in sample extraction and analysis. Duplicate sample RPDs can vary substantially between soil samples dependent on the inherent soil heterogeneity and matrix complexity. RPDs less than about 50% are considered good for soil samples. Table A-5 shows the laboratory duplicate results.
Surrogate compounds were added to each sample prior to extraction to monitor the analytical performance. The surrogate compounds used for this study included nitrobenzene-d5, 2-fluorobiphenyl, p-terphenyl-d14, and benzo(a)pyrene-d12. The recoveries of surrogate compounds were within the QAPP-specified limits of 50 to 150% with few exceptions.
In addition to surrogate compound spikes, selected samples were spiked with all 17 PAHs at known concentrations. The recoveries of matrix-spiked compounds were within the QAPP-specified limits of 50 to 150% with few exceptions.
4-1
4 CONCLUSIONS
As part of the ongoing EPRI nationwide background PAH study, two municipalities in western New York with a population of at least 10,000 people and a population density of at least 1,000 people per square mile were sampled. Samples were collected from ten sites in Canandaigua and 20 sites in Rochester, with two locations sampled at each site. At each location, samples were collected from 0 to 2.54 cm and 2.54 to 15 cm below grade, after removing surface vegetation, if present. During sampling, subjective classifications of the site use and area use were made along with detailed descriptions of the soil types contained in each sample. A total of 30 sites were sampled during this part of the study.
The samples that were collected in western New York were composited in the laboratory, with one composite sample generated for each site. Specifically, the four discrete samples collected from each site were composited to create one sample that represented 0 to 15 cm for that site.
The soil characteristics were representative of typical surface soils found in many areas of the United States. The samples contained various amounts of silt, sand, clay, and gravel. The median TOC content was 3.1% and the median percent solids content was 90%.
Summary statistics were performed on the data set, with all samples expressed as a 0 to 15 cm sample interval. The results of these manipulations indicate that the data are generally lognormally distributed for all of the PAHs. As shown in Table 3-1, benzo(a)pyrene concentrations ranged from 7.5 to 4,740 µg/kg with a median concentration of 431 µg/kg. The upper quartile concentration was 1,040 µg/kg, while the upper 95th percentile concentration was 2,770 µg/kg.
These concentrations appear to be consistent with literature-reported levels for anthropogenic background in small to medium-sized residential, commercial, and light industrial areas, but notably lower than background observed in more highly populated cities and commercial/industrial areas, as shown in Table 4-1 (Bradley et al., 1994 and MA DEP, 2002). In comparison to the previous EPRI study results, these benzo(a)pyrene concentrations were generally higher, as illustrated by the mean, median, and 95th percentile values, although the maximum concentration measured was lower.
Conclusions
4-2
Table 4-1 Summary Benzo(a)pyrene Statistics for Other Data Sets Compared to this Study (µg/kg)
Data Set N Range Mean Median 95th Percentile
CA/T Project1 873 31 – 230,000 NA 300 17,000
LSPA Project2 489 ND – 222,000 NA 440 NA
Watertown3 17 600 – 6,080 NA NA 4,770
Med City/Mill Brook4 67 ND – 9,700 NA NA 3,300
ENSR – Urban Soils5 62 ND – 13,000 1,320 NA NA
EPRI Study – 20026 218 ND – 11,700 382 98.2 1,510
EPRI Study – Western NY Data
30 7.5 – 4,740 830 431 2,770
1 Data collected by the Massachusetts Highway Department as part of the Central Artery/Tunnel (CA/T) project in Boston, presented in the draft document “CA/T ROW Background Soil Contaminant Assessment”, Camp Dresser and McKee, April 1996. 2 Preliminary data submitted to the Massachusetts Licensed Site Professional Association by its members and compiled in the personal communication, “Summary of Selected Results, LSPA Anthropogenic Fill Soils Projects”, LSPA, April 2001. 3 Site-specific background samples collected in Watertown, MA. 4 Site-specific background samples collected in Worcester, MA. 5 Data from background surficial samples collected in Boston, MA, Providence, RI, and Springfield, MA by ENSR (published in Bradley et al., 1994). 6 EPRI background PAH study data from New York and Illinois reported in Technical Update 1005295 (EPRI, 2002).
5-1
5 REFERENCES
Blumer, M., “Polycyclic Aromatic Compounds in Nature,” Scientific American, Vol. 243, pp. 35-45 (1976).
Bradley, L.J.N., B.H. Magee, and S.L. Allen, “Background Levels of Polycyclic Aromatic Hydrocarbons (PAHs) and Selected Metals in New England Urban Soils”. Journal of Soil Contamination. Vol. 3, pp. 349-361 (1994).
Gilbert, R. O., 1987. Statistical Methods for Environmental Pollution Monitoring. John Wiley & Sons, Inc., New York, NY.
Literature Review of Polycyclic Aromatic Hydrocarbons, EPRI, Palo Alto, CA: 2000. TR-114755.
Massachusetts Department of Environmental Protection (MA DEP). May 2002. Technical Update – Background Levels of Polycyclic Aromatic Hydrocarbons and Metals in Soil.
Polycyclic Aromatic Hydrocarbons (PAHs) in Surface Soil, EPRI, Palo Alto, CA: 2002. 105295.
90-3, Standard Methods of Chemical Analysis, 6th Edition, D. Van Nostrand Co., Princeton, NJ, 1963.
Statistica version 6.0, StatSoft, Inc., Tulsa OK 74104.
U.S. EPA, 1986. Test Methods for Evaluating Solid Waste Physical/Chemical Methods. SW-846 Third Edition. Office of Solid Waste, U.S. EPA, Washington, D.C.
A-1
A TABLES
Table A-1 Sample Soil Descriptions, Total Organic Carbon, and Percent Solids
Field ID Sample Description TOC (mg/kg) % Solids
NYCA01-Comp(0-15 Silt with sand 37,300 82.5%NYCA02-Comp(0-15 Silt with gravel and clay 22,800 80.8%NYCA03-Comp(0-15 Silt with sand and gravel 63,100 86.9%NYCA04-Comp(0-15 Silt with sand and gravel 29,600 83.1%NYCA05-Comp(0-15 Sand with silt 6,010 90.9%NYCA06-Comp(0-15 Silt with sand 39,300 81.2%NYCA07-Comp(0-15 Silt with clay and gravel 31,300 85.6%NYCA08-Comp(0-15 Silt with sand and clay 30,800 84.5%NYCA09-Comp(0-15 Silt with sand 25,900 85.2%NYCA10-Comp(0-15 Silt with sand and gravel 28,300 85.4%NYRO01-Comp(0-15 Silt with sand and gravel 31,700 90.6%NYRO02-Comp(0-15 Silt with clay and gravel 23,900 88.4%NYRO03-Comp(0-15 Sand with silt 23,100 94.6%NYRO04-Comp(0-15 Silt with sand and gravel 14,400 88.2%NYRO05-Comp(0-15 Silt with clay and gravel 33,700 94.2%NYRO06-Comp(0-15 Silt with sand and clay 48,700 92.2%NYRO07-Comp(0-15 Silt with sand and gravel 101,000 92.6%NYRO08-Comp(0-15 Silt with sand and clay 31,400 92.5%NYRO09-Comp(0-15 Silt with sand and gravel 32,400 91.5%NYRO10-Comp(0-15 Sand with silt and gravel 23,100 95.1%NYRO11-Comp(0-15 Silt with sand and clay 22,300 92.8%NYRO12-Comp(0-15 Silt with gravel and sand 24,500 91.3%NYRO13-Comp(0-15 Silt with sand and gravel 22,600 89.6%NYRO14-Comp(0-15 Silt with sand and clay 45,800 85.1%NYRO15-Comp(0-15 Sand with silt and gravel 30,900 92.5%NYRO16-Comp(0-15 Sand with silt and clay 18,800 94.4%NYRO17-Comp(0-15 Silt with sand and gravel 31,100 88.5%NYRO18-Comp(0-15 Silt with sand and gravel 35,400 92.0%NYRO19-Comp(0-15 Silt with sand and gravel 38,600 90.4%NYRO20-Comp(0-15 Silt with sand and gravel 30,000 88.5%
Tables
A-2
Table A-2 Key to Abbreviations for PAHs
Compounds Abbreviation
Naphthalene NAP
2-Methylnaphthalene MN2
Acenaphthylene ACY
Acenaphthene ACE
Fluorene FLU
Phenanthrene PHE
Anthracene ANT
Fluoranthene FLR
Pyrene PYR
Benz(a)anthracene BAA
Chrysene CHR
Benzo(b)fluoranthene BBF
Benzo(k)fluoranthene BKF
Benzo(a)pyrene BAP
Indeno(1,2,3-cd)pyrene IND
Dibenz(a,h)anthracene DAA
Benzo(g,h,i)perylene BGP
Tabl
es
A-3
Tabl
e A
-3
Pol
ycyc
lic A
rom
atic
Hyd
roca
rbon
(PA
H) R
esul
ts in
Bac
kgro
und
Sur
face
Soi
ls
Site
N
YC
A01
N
YC
A02
NY
CA
03N
YC
A04
NY
CA
05N
YC
A06
NY
CA
07N
YC
A08
Sam
ple
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Uni
ts
µg/K
g µg
/Kg
µg/K
g µg
/Kg
µg/K
g µg
/Kg
µg/K
g µg
/Kg
Nap
htha
lene
21
.5
20
.6
U89
.9
11
4
8.01
J
98.7
46.3
276
2-M
ethy
lnap
htha
lene
4.
47
J20
.6
U92
.1
38
.9
2.
51
J 57
.2
22
.4
26
5
Ace
naph
thyl
ene
6.08
J
20.6
U
417
51
4
4.06
J
76.5
89.8
128
Ace
naph
then
e 7.
23
J20
.6
U31
.2
50
.7
18
.3
U29
.9
10
.3
J16
.5
J
Fluo
rene
8.
02
J20
.6
U41
.8
88
.3
18
.3
U40
.5
25
.2
28
.7
Phe
nant
hren
e 15
9
6.48
J
1,07
0
1,55
0
14.9
J
664
31
6
546
Ant
hrac
ene
30
20
.6
U29
7
583
1.
37
J 14
2
97.3
106
Fluo
rant
hene
41
8
19.5
J
3,01
0
3,55
0
27.6
1,21
0
732
1,
310
Pyr
ene
320
17
J
2,49
0
2,91
0
24.1
962
52
2
1,01
0
Ben
z(a)
anth
race
ne
131
9.
01
J 1,
420
2,
100
8.
44
J 49
3
335
53
8
Chr
ysen
e 17
3
11.6
J
1,82
0
2,12
0
18.3
616
38
3
621
Ben
zo(b
)flu
oran
then
e 11
6
5.3
J 1,
770
2,
370
10
.3
J 48
0
316
53
9
Ben
zo(k
)flu
oran
then
e 16
2
9.36
J
1,37
0
1,76
0
13.7
J
466
42
0
549
Ben
zo(a
)pyr
ene
164
7.
48
J 1,
770
2,
770
10
.6
J 57
2
417
64
7
Inde
no(1
,2,3
-84
.5
5.
03
J 1,
240
2,
170
8.
17
J 33
1
340
34
4
Dib
enz(
a,h)
anth
race
ne
27.3
J
1.83
J
476
81
9
4.93
J
130
11
4
129
Ben
zo(g
,h,i)
pery
lene
10
2
7.28
J
1,36
0
2,42
0
12.9
J
370
36
7
356
D
Con
cent
ratio
n ob
tain
ed f
rom
a d
ilute
d ex
trac
t E
E
stim
ated
con
cent
ratio
n de
tect
ed a
bove
cal
ibra
tion
limit
J
Est
imat
ed v
alue
det
ecte
d be
low
the
quan
titat
ion
limit
but a
bove
the
dete
ctio
n lim
it U
T
arge
t com
poun
d no
t det
ecte
d at
or
abov
e th
e gi
ven
conc
entr
atio
n
Ta
bles
A-4
Tabl
e A
-3 (c
ont.)
P
olyc
yclic
Aro
mat
ic H
ydro
carb
on (P
AH
) Res
ults
in B
ackg
roun
d S
urfa
ce S
oils
Site
N
YC
A09
N
YC
A10
NY
RO
01N
YR
O02
NY
RO
03N
YR
O04
NY
RO
05N
YR
O06
Sam
ple
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Uni
ts
µg/K
g µg
/Kg
µg/K
g µg
/Kg
µg/K
g µg
/Kg
µg/K
g µg
/Kg
Nap
htha
lene
38
.8
34
.7
13
3
32.1
90.8
27.5
443
88
2-M
ethy
lnap
htha
lene
18
.3
J11
.9
J10
1
13.9
J
41.9
20.7
360
64
Ace
naph
thyl
ene
174
67
.7
30
6
7.12
J
222
37
.8
71
3
91.9
Ace
naph
then
e 10
.5
J5.
14
J20
9
3.78
J
51.8
4.07
260
14
.4
J
Fluo
rene
7.
08
J15
.3
J20
9
3.93
J
51
7
31
9
28.3
Phe
nant
hren
e 16
8
239
2,
820
77
.3
97
1
118
4,
820
D40
1
Ant
hrac
ene
86.2
70.7
579
12
.9
J30
8
30.6
1,33
0
73.5
Fluo
rant
hene
57
7
706
4,
970
D24
0
2,76
0
354
9,
010
D72
4
Pyr
ene
513
55
1
4,09
0 D
199
2,
280
33
8
6,75
0 D
604
Ben
z(a)
anth
race
ne
399
30
8
2,25
0
92.8
1,86
0
179
3,
910
D32
4
Chr
ysen
e 43
7
306
2,
490
12
9
2,09
0
218
4,
330
D39
8
Ben
zo(b
)flu
oran
then
e 46
6
210
2,
240
11
1
1,99
0
239
3,
340
D39
9
Ben
zo(k
)flu
oran
then
e 41
8
282
98
8
106
1,
440
23
0
3,35
0 D
347
Ben
zo(a
)pyr
ene
622
32
1
2,13
0
135
2,
360
21
8
4,74
0 D
444
Inde
no(1
,2,3
-46
0
150
1,
700
81
.3
1,
850
10
4
2,33
0 D
251
Dib
enz(
a,h)
anth
race
ne
187
51
.8
64
8
29.1
J
552
39
.4
81
8 D
89
Ben
zo(g
,h,i)
pery
lene
54
2
143
1,
880
90
.4
2,
010
95
.8
2,
100
D26
9
D
Con
cent
ratio
n ob
tain
ed f
rom
a d
ilute
d ex
trac
t E
E
stim
ated
con
cent
ratio
n de
tect
ed a
bove
cal
ibra
tion
limit
J
Est
imat
ed v
alue
det
ecte
d be
low
the
quan
titat
ion
limit
but a
bove
the
dete
ctio
n lim
it U
T
arge
t com
poun
d no
t det
ecte
d at
or
abov
e th
e gi
ven
conc
entr
atio
n
Tabl
es
A-5
Tabl
e A
-3 (c
ont.)
P
olyc
yclic
Aro
mat
ic H
ydro
carb
on (P
AH
) Res
ults
in B
ackg
roun
d S
urfa
ce S
oils
Site
N
YR
O07
N
YR
O08
NY
RO
09N
YR
O10
NY
RO
11N
YR
O12
NY
RO
13N
YR
O14
Sam
ple
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Uni
ts
µg/K
g µg
/Kg
µg/K
g µg
/Kg
µg/K
g µg
/Kg
µg/K
g µg
/Kg
Nap
htha
lene
66
7
149
12
2
74.9
36.6
17.5
J
16.2
J
12.9
J
2-M
ethy
lnap
htha
lene
81
6
68
15
5
37.1
37.2
13.8
J
7 J
9.26
J
Ace
naph
thyl
ene
287
32
.6
36
.7
72
.8
17
.1
J13
.8
J7.
58
J 8.
36
J
Ace
naph
then
e 80
.3
23
1
16.5
107
17
.7
J7.
2 J
18.6
U
4.54
J
Fluo
rene
79
.4
25
4
26.1
111
24
.2
9.
72
J18
.6
U4.
83
J
Phe
nant
hren
e 1,
550
2,
570
39
0
1,66
0
313
14
8
68.9
96
Ant
hrac
ene
395
59
2
81.4
378
58
.9
28
.6
11
.4
J 19
.6
J
Fluo
rant
hene
1,
580
3,
040
64
8
1,55
0
588
33
4
117
23
6
Pyr
ene
1,22
0
2,38
0
555
96
6
473
25
9
95.8
199
Ben
z(a)
anth
race
ne
983
1,
040
26
5
864
20
6
119
53
.2
83
.3
Chr
ysen
e 1,
190
1,
120
29
9
930
26
2
169
77
.2
12
7
Ben
zo(b
)flu
oran
then
e 79
1
968
30
5
697
27
2
135
55
.3
10
1
Ben
zo(k
)flu
oran
then
e 64
5
749
28
0
659
24
2
155
59
.4
10
5
Ben
zo(a
)pyr
ene
1,04
0
1,07
0
267
98
2
286
16
7
76.2
130
Inde
no(1
,2,3
-81
4
479
11
8
693
15
6
94.1
13.7
J
77.1
Dib
enz(
a,h)
anth
race
ne
306
18
6
52.9
238
58
.1
25
.5
J2.
3 J
27.7
J
Ben
zo(g
,h,i)
pery
lene
96
6
524
11
4
822
18
7
109
31
.2
98
.4
D
Con
cent
ratio
n ob
tain
ed f
rom
a d
ilute
d ex
trac
t E
E
stim
ated
con
cent
ratio
n de
tect
ed a
bove
cal
ibra
tion
limit
J
Est
imat
ed v
alue
det
ecte
d be
low
the
quan
titat
ion
limit
but a
bove
the
dete
ctio
n lim
it U
T
arge
t com
poun
d no
t det
ecte
d at
or
abov
e th
e gi
ven
conc
entr
atio
n
Ta
bles
A-6
Tabl
e A
-3 (c
ont.)
P
olyc
yclic
Aro
mat
ic H
ydro
carb
on (P
AH
) Res
ults
in B
ackg
roun
d S
urfa
ce S
oils
Site
N
YR
O15
N
YR
O16
NY
RO
17N
YR
O18
NY
RO
19N
YR
O20
Sam
ple
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Com
p(0-
15)
Uni
ts
µg/K
g µg
/Kg
µg/K
g µg
/Kg
µg/K
g µg
/Kg
Nap
htha
lene
33
.4
25.1
16.8
J11
.7J
221
47.6
2-M
ethy
lnap
htha
lene
12
.5
J9.
17
J8.
79
J2.
25
J35
6
38.4
Ace
naph
thyl
ene
73.6
5.17
J
6.81
J
3.18
J
103
35
.4
Ace
naph
then
e 16
.6
J2.
67
J4.
34
J5.
53
J88
.4
97
.9
Fluo
rene
18
.5
1.
91
J4.
1 J
5.26
J
104
11
5
Phe
nant
hren
e 65
0
82.4
92.8
107
1,
190
1,
060
Ant
hrac
ene
117
13
.1
J18
.4
J20
.3
35
3
317
Fluo
rant
hene
1,
580
15
1
175
20
6
2,47
0
1,76
0
Pyr
ene
1,21
0
116
13
4
155
1,
980
1,
450
Ben
z(a)
anth
race
ne
561
63
.8
75
95.2
1,14
0
801
Chr
ysen
e 79
2
106
11
1
124
1,
420
94
3
Ben
zo(b
)flu
oran
then
e 70
1
57.4
56.5
84.2
1,89
0
1,11
0
Ben
zo(k
)flu
oran
then
e 51
4
82.9
94.5
89.2
1,64
0
1,15
0
Ben
zo(a
)pyr
ene
798
88
.1
10
7
124
1,
520
92
3
Inde
no(1
,2,3
-33
9
30.2
J
31.6
J
50
48
9
240
Dib
enz(
a,h)
anth
race
ne
105
8.
53
J12
.7
J18
.3
J18
5
88.6
Ben
zo(g
,h,i)
pery
lene
42
6
43.5
39.8
62.4
448
22
5
D
Con
cent
ratio
n ob
tain
ed f
rom
a d
ilute
d ex
trac
t E
E
stim
ated
con
cent
ratio
n de
tect
ed a
bove
cal
ibra
tion
limit
J
Est
imat
ed v
alue
det
ecte
d be
low
the
quan
titat
ion
limit
but a
bove
the
dete
ctio
n lim
it U
T
arge
t com
poun
d no
t det
ecte
d at
or
abov
e th
e gi
ven
conc
entr
atio
n
Tables
A-7
Table A-4 Field Duplicate Results with Relative Percent Differences (RPDs)
Site NYRO04 NYRO04 NYRO09 NYRO09 Sample Comp(0-15) Comp(0-15)D RPD Comp(0-15) Comp(0-15)D RPD
Units µg/Kg µg/Kg µg/Kg µg/Kg
Naphthalene 27.5 36.3 28 122 161 28
2-Methylnaphthalene 20.7 31.2 40 155 211 31
Acenaphthylene 37.8 52.8 33 36.7 40.7 10
Acenaphthene 4.07 4.14 2 16.5 23.9 37
Fluorene 7 8.49 19 26.1 35.4 30
Phenanthrene 118 177 40 390 464 17
Anthracene 30.6 38.6 23 81.4 92.7 13
Fluoranthene 354 523 39 648 721 11
Pyrene 338 463 31 555 633 13
Benz(a)anthracene 179 223 22 265 286 8
Chrysene 218 272 22 299 349 15
Benzo(b)fluoranthene 239 284 17 305 313 3
Benzo(k)fluoranthene 230 297 25 280 316 12
Benzo(a)pyrene 218 245 12 267 295 10
Indeno(1,2,3- 104 112 7 118 122 3
Dibenz(a,h)anthracene 39.4 41.8 6 52.9 55.6 5
Benzo(g,h,i)perylene 95.8 101 5 114 124 8
D Concentration obtained from a diluted extract E Estimated concentration detected above calibration limit J Estimated value detected below the quantitation limit but above the detection limit U Target compound not detected at or above the given concentration RPD Relative percent difference, equal to the difference divided by the average of the numbers multiplied by 100
Tables
A-8
Table A-5 Laboratory Duplicate Results with Relative Percent Differences (RPDs)
Site NYCA03 NYCA03 NYRO10 NYRO10 Sample Comp(0-15) Comp(0-15)R RPD Comp(0-15) Comp(0-15)R RPD
Units µg/Kg µg/Kg µg/Kg µg/Kg
Naphthalene 89.9 127 34 74.9 60.4 21
2-Methylnaphthalene 92.1 106 14 37.1 33.5 10
Acenaphthylene 417 1,060 87 72.8 40.1 58
Acenaphthene 31.2 81.4 89 107 74.2 36
Fluorene 41.8 163 118 111 71.7 43
Phenanthrene 1,070 2,550 82 1,660 1,070 43
Anthracene 297 1,080 114 378 221 52
Fluoranthene 3,010 8,080 D 91 1,550 939 49
Pyrene 2,490 5,900 D 81 966 647 40
Benz(a)anthracene 1,420 4,230 D 99 864 562 42
Chrysene 1,820 4,460 D 84 930 647 36
Benzo(b)fluoranthene 1,770 3,340 D 61 697 461 41
Benzo(k)fluoranthene 1,370 4,290 D 103 659 496 28
Benzo(a)pyrene 1,770 4,760 D 92 982 670 38
Indeno(1,2,3- 1,240 3,120 86 693 518 29
Dibenz(a,h)anthracene 476 1,210 87 238 148 47
Benzo(g,h,i)perylene 1,360 3,350 85 822 596 32
D Concentration obtained from a diluted extract E Estimated concentration detected above calibration limit J Estimated value detected below the quantitation limit but above the detection limit U Target compound not detected at or above the given concentration RPD Relative percent difference, equal to the difference divided by the average of the numbers multiplied by 100
© 2003 Electric Power Research Institute (EPRI), Inc.All rightsreserved. Electric Power Research Institute and EPRI are registered service marks of the Electric Power Research Institute, Inc.EPRI. ELECTRIFY THE WORLD is a service mark of the ElectricPower Research Institute, Inc.
Printed on recycled paper in the United States of America
1005296
Program:
MGP Site Management
EPRI • 3412 Hillview Avenue, Palo Alto, California 94304 • PO Box 10412, Palo Alto, California 94303 • USA800.313.3774 • 650.855.2121 • [email protected] • www.epri.com
About EPRI
EPRI creates science and technology solutions for
the global energy and energy services industry. U.S.
electric utilities established the Electric Power
Research Institute in 1973 as a nonprofit research
consortium for the benefit of utility members, their
customers, and society. Now known simply as EPRI,
the company provides a wide range of innovative
products and services to more than 1000 energy-
related organizations in 40 countries. EPRI’s
multidisciplinary team of scientists and engineers
draws on a worldwide network of technical and
business expertise to help solve today’s toughest
energy and environmental problems.
EPRI. Electrify the World